Notes
Start with the area you care about, then scan the latest notes inside it.
Start here
Choose where to begin. OpenClaw is the broadest entry point.
Source-code teardown, subsystem support notes, and the operator setup path.
The start-here tutorial lane for learning how to build an AI agent, with the first working agent tutorial first, the framework route second, and support notes after the route is chosen.
Stack decisions, CLI workflows, and the tooling layer around production agent systems.
Browse notes
Main reads come first, with related notes close behind.
Source-code teardown, subsystem support notes, and the operator setup path.
Deep dive into OpenClaw's heartbeat and cron systems — the architecture that turns a reactive chatbot into an autonomous AI agent that wakes itself, schedules its own future, and improves while you sleep.
Deep dive into OpenClaw gateway-first architecture — how a single WebSocket, channel plugin system, and block streaming engine let one AI agent show up everywhere.
Deep dive into how OpenClaw agents modify their own personality files, create new skills, and drift into emergent behaviors — the architecture of AI self-modification.
Copyable OpenClaw `HEARTBEAT.md` templates and the rules for quiet hours, `HEARTBEAT_OK`, and keeping heartbeat silent by default.
Skimmable OpenClaw system map covering the gateway, sessions, memory files, heartbeat and cron, and the execution layer from inbound message to action.
Practical OpenClaw scheduling guide: when heartbeat should batch recurring checks, when cron should own exact reminders, and how to avoid notification noise.
The start-here tutorial lane for learning how to build an AI agent, with the first working agent tutorial first, the framework route second, and support notes after the route is chosen.
Acceleration comes from shipping the smallest working system, then compounding it with tight feedback loops.
Most AI agent architecture still imitates human teams. The better model is factories: queues, workcells, QA gates, and auditable async worker systems.
A practical field guide to running coding agents safely: scope, isolation, verification, and review.
Hermes Agent review for builders: what the repo actually does, what we validated, what is hype, and which patterns are worth stealing.
Stack decisions, CLI workflows, and the tooling layer around production agent systems.
I built datafast-cli and pointed an autonomous AI agent at it. 13 commands, 2 bugs found, and the 5 principles that make CLI tools genuinely useful as AI agent tools. If you're still choosing the broader first-agent path before the tooling layer, start at /build-ai-agent.
CLI-first SEO becomes one of the most practical ai developer tools when keywords, SERPs, audits, and ranks turn into machine-readable handoffs.
A practical field guide to the layer above the coding agent: when to use native CLIs, when wrappers help, and when a full harness is worth the complexity.
A practical field guide to cross-agent handoff: what belongs in the packet, when to resume instead of switch, and how to move work between coding agents without turning the workflow into mush.
A practical guide to coding agent wrappers: where they help, where they degrade workflow quality, and how to judge native CLI vs wrapper vs harness without policy melodrama.
ClawSweeper is an AI repo-maintenance worker with typed decisions, durable artifacts, and a proposal/apply split. Here’s what it actually does and why the design matters.
OpenAI Symphony is an issue-driven coding-agent orchestrator with repo-owned workflow contracts, reconciliation loops, and per-issue workspaces. Here’s what it actually does and what builders should steal.
Additional notes from the library.
Claude Managed Agents gives Anthropic a hosted agent runtime with sessions, environments, and tools. Here’s what it actually ships, what it gets right, and what control you give up.
A practical OpenClaw Codex guide for running Codex and Claude Code through ACP, with the real runtime boundary, thread workflow, permission caveats, and decision rule.
Karpathy’s autoresearch is a real autonomous experiment loop, but it is much narrower than the hype suggests. Here is what the repo actually does, what breaks when you generalize it, and the one pattern worth stealing.
A deep, command-level teardown of claudeagentsdk (#005): an open-source agent workspace built around the Anthropic Agent SDK, with a FastAPI backend, a Vite/React frontend, and an optional Vercel Sandbox runner for async, reproducible runs.
A command-level teardown of the Starkslab inbox-to-execution loop: intake, triage, routing, artifact discipline, incidents, handoffs, metrics, and checklist controls.
A command-level, evidence-first teardown of where OpenClaw fits in an ai developer tools stack: architecture, workflows, incidents, throughput, and adoption boundaries.
I built trustmrr-cli — a TypeScript CLI giving AI agents access to verified revenue data for 4,900+ startups. Here's the architecture, the API workarounds, and why agent-native CLI tools are the missing layer.
I built an X post scheduler from scratch — Express, Postgres, cron — and had an AI coding agent write most of it. Here's the architecture, the deployment, and why simple AI agent automation beats over-engineering.